.png){kind=icon}
.png){kind=icon}
Return
Constructing Composite Indicators Through Extreme Values Reductions-Ordered Weighted Averaging: Human Development Index
.png){kind=icon}
OA
AI
DOI：10.1109/ACCESS.2025.3546529.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The Human Development Index is a complex social phenomenon composed of multiple criteria. While many studies are concerned with dealing with outliers in the representation of this composite indicator, the problem of extreme values is overlooked. This study falls within the multi-criteria decision framework aimed at introducing a new approach to constructing composite indicators specifically that addresses the issue of extreme values, called the Extreme Values Reductions-Ordered Weighted Averaging approach. The study findings reveal that popular methods such as Principal Component Analysis, Benefit of the Doubt, Entropy, and Equal Weights ignore outliers and bias the composite indicator scores. The Extreme Value Reduction approach provides an unbiased representation of the Human Development Index, preventing extreme lower and higher values from uncovering weaknesses and unmasking strengths of the decision-making units. The possibility of adjusting the weights of the intermediate values also provides the opportunity to reveal nuances that are difficult to detect by controlling the coefficient of variation of the composite indicator scores.
Keywords:
Indexes
Decision making
Entropy
Systematic literature review
Principal component analysis
Hands
Weighted sum model
Logic
Weight measurement
Urban areas
Multicriteria methods
multidimensional problems
composite indicators
ordered weighted averaging
extreme values reductions
human development index
Journal
IF:
3.6
Papers:
9.8W
Citations:
29.4W
View more organizations